Daily reservoir inflow forecasting using multiscale deep feature learning with hybrid models

Summary Inflow forecasting applies data supports for the operations and managements of reservoirs. A multiscale deep feature learning (MDFL) method with hybrid models is proposed in this paper to deal with the daily reservoir inflow forecasting. Ensemble empirical mode decomposition and Fourier spectrum are first employed to extract multiscale (trend, period and random) features, which are then represented by three deep belief networks (DBNs), respectively. The weights of each DBN are subsequently applied to initialize a neural network (D-NN). The outputs of the three-scale D-NNs are finally reconstructed using a sum-up strategy toward the forecasting results. A historical daily inflow series (from 1/1/2000 to 31/12/2012) of the Three Gorges reservoir, China, is investigated by the proposed MDFL with hybrid models. For comparison, four peer models are adopted for the same task. The results show that, the present model overwhelms all the peer models in terms of mean absolute percentage error (MAPE = 11.2896%), normalized root-mean-square error (NRMSE = 0.2292), determination coefficient criteria ( R 2  = 0.8905), and peak percent threshold statistics (PPTS(5) = 10.0229%). The addressed method integrates the deep framework with multiscale and hybrid observations, and therefore being good at exploring sophisticated natures in the reservoir inflow forecasting.

[1]  Yong-Huang Lin,et al.  Applying fuzzy grey modification model on inflow forecasting , 2012, Eng. Appl. Artif. Intell..

[2]  Yee Whye Teh,et al.  Rate-coded Restricted Boltzmann Machines for Face Recognition , 2000, NIPS.

[3]  Robert J. Abrahart,et al.  Including spatial distribution in a data‐driven rainfall‐runoff model to improve reservoir inflow forecasting in Taiwan , 2014 .

[4]  H. Abarbanel,et al.  Determining embedding dimension for phase-space reconstruction using a geometrical construction. , 1992, Physical review. A, Atomic, molecular, and optical physics.

[5]  Mac McKee,et al.  Multivariate Bayesian Regression Approach to Forecast Releases from a System of Multiple Reservoirs , 2011 .

[6]  R. Khosa,et al.  Wavelets-based non-linear model for real-time daily flow forecasting in Krishna River , 2013 .

[7]  Lin Qiu,et al.  Support Vector Machine with Particle Swarm Optimization for Reservoir Annual Inflow Forecasting , 2010, 2010 International Conference on Artificial Intelligence and Computational Intelligence.

[8]  A. K. Lohani,et al.  Hydrological time series modeling: A comparison between adaptive neuro-fuzzy, neural network and autoregressive techniques , 2012 .

[9]  Pu Wang,et al.  Additive Model for Monthly Reservoir Inflow Forecast , 2015 .

[10]  Chuan Li,et al.  Criterion fusion for spectral segmentation and its application to optimal demodulation of bearing vibration signals , 2015 .

[11]  Fraser,et al.  Independent coordinates for strange attractors from mutual information. , 1986, Physical review. A, General physics.

[12]  Mawada E. Abdellatif,et al.  Comparison of artificial neural networks and autoregressive model for inflows forecasting of Roseires Reservoir for better prediction of irrigation water supply in Sudan , 2015 .

[13]  N. Huang,et al.  A study of the characteristics of white noise using the empirical mode decomposition method , 2004, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[14]  Pascal Vincent,et al.  The Difficulty of Training Deep Architectures and the Effect of Unsupervised Pre-Training , 2009, AISTATS.

[15]  Norden E. Huang,et al.  Ensemble Empirical Mode Decomposition: a Noise-Assisted Data Analysis Method , 2009, Adv. Data Sci. Adapt. Anal..

[16]  Xiao-Lei Zhang,et al.  Deep Belief Networks Based Voice Activity Detection , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[17]  Ming Liang,et al.  Extraction of oil debris signature using integral enhanced empirical mode decomposition and correlated reconstruction , 2011 .

[18]  M. Valipour,et al.  Parameters Estimate of Autoregressive Moving Average and Autoregressive Integrated Moving Average Models and Compare Their Ability for Inflow Forecasting , 2012 .

[19]  Vahid Nourani,et al.  A Multivariate ANN-Wavelet Approach for Rainfall–Runoff Modeling , 2009 .

[20]  Chuan Li,et al.  A generalized synchrosqueezing transform for enhancing signal time-frequency representation , 2012, Signal Process..

[21]  Mohammad H. Aminfar,et al.  A combined neural-wavelet model for prediction of Ligvanchai watershed precipitation , 2009, Eng. Appl. Artif. Intell..

[22]  Vahid Nourani,et al.  Development of Nonlinear Model Based on Wavelet-ANFIS for Rainfall Forecasting at Klang Gates Dam , 2014, Water Resources Management.

[23]  Tapani Raiko,et al.  Improved Learning of Gaussian-Bernoulli Restricted Boltzmann Machines , 2011, ICANN.

[24]  N. K. Goel,et al.  Improving real time flood forecasting using fuzzy inference system , 2014 .

[25]  Amy Loutfi,et al.  A review of unsupervised feature learning and deep learning for time-series modeling , 2014, Pattern Recognit. Lett..

[26]  Yung-Cheol Byun,et al.  An Adaptive Stopping Criterion for Backpropagation Learning in Feedforward Neural Network , 2014, MUE 2014.

[27]  Henry D. I. Abarbanel,et al.  Analysis of Observed Chaotic Data , 1995 .

[28]  Zhongbo Yu,et al.  Real-time flood forecast using the coupling support vector machine and data assimilation method , 2014 .

[29]  Aini Hussain,et al.  Erratum to: Daily Forecasting of Dam Water Levels: Comparing a Support Vector Machine (SVM) Model With Adaptive Neuro Fuzzy Inference System (ANFIS) , 2013, Water Resources Management.

[30]  Florentino Fernández Riverola,et al.  A hybrid artificial intelligence model for river flow forecasting , 2013, Appl. Soft Comput..

[31]  B. De Schutter,et al.  Distributed tree-based model predictive control on a drainage water system , 2013 .

[32]  Wei Xu,et al.  A two stage Bayesian stochastic optimization model for cascaded hydropower systems considering varying uncertainty of flow forecasts , 2014 .

[33]  Y. Osman,et al.  Comparison of Artificial Neural Networks and Autoregressive Model to Forecast Inflows to Roseires Reservoir for better Prediction of Irrigation Water Supply in the Sudan , 2015 .

[34]  Wenhao Huang,et al.  Deep Architecture for Traffic Flow Prediction: Deep Belief Networks With Multitask Learning , 2014, IEEE Transactions on Intelligent Transportation Systems.

[35]  Ronny Berndtsson,et al.  Advances in Data-Based Approaches for Hydrologic Modeling and Forecasting , 2010 .

[36]  Vahid Nourani.  Reply to comment on ‘Nourani V, Mogaddam AA, Nadiri AO. 2008. An ANN-based model for spatiotemporal groundwater level forecasting. Hydrological Processes 22: 5054–5066’ , 2009 .

[37]  J. Ross Macdonald.  Review and Reveal the Reviewers , 1996 .

[38]  Diego Cabrera,et al.  Multimodal deep support vector classification with homologous features and its application to gearbox fault diagnosis , 2015, Neurocomputing.

[39]  Tianrui Li,et al.  Learning features from High Speed Train vibration signals with Deep Belief Networks , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[40]  Qiongji Jin,et al.  Design of Deep Belief Networks for Short-Term Prediction of Drought Index Using Data in the Huaihe River Basin , 2012 .

[41]  K. Budu,et al.  Comparison of Wavelet-Based ANN and Regression Models for Reservoir Inflow Forecasting , 2014 .

[42]  Kwok-wing Chau,et al.  Improving Forecasting Accuracy of Annual Runoff Time Series Using ARIMA Based on EEMD Decomposition , 2015, Water Resources Management.

[43]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[44]  Kazuyuki Murase,et al.  Bangla handwritten character recognition using deep belief network , 2014, 2013 International Conference on Electrical Information and Communication Technology (EICT).

[45]  Umut Okkan,et al.  The combined use of wavelet transform and black box models in reservoir inflow modeling , 2013 .

[46]  Joong Hoon Kim,et al.  Forecasting performance of LS-SVM for nonlinear hydrological time series , 2012 .

[47]  M. Valipour,et al.  Comparison of the ARMA, ARIMA, and the autoregressive artificial neural network models in forecasting the monthly inflow of Dez dam reservoir , 2013 .

[48]  Gwo-Fong Lin,et al.  An RBF network with a two-step learning algorithm for developing a reservoir inflow forecasting model , 2011 .

[49]  Yoshua Bengio,et al.  Greedy Layer-Wise Training of Deep Networks , 2006, NIPS.

[50]  V. Jothiprakash,et al.  Effect of Pruning and Smoothing while Using M5 Model Tree Technique for Reservoir Inflow Prediction , 2011 .

[51]  Hongjuan Zeng,et al.  A Distributed Hydrological Model Driven by Multi-Source Spatial Data and Its Application in the Ili River Basin of Central Asia , 2014, Water Resources Management.

[52]  Pascal Vincent,et al.  Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives , 2012, ArXiv.

[53]  M. U. Kale,et al.  Reservoir Inflow Forecasting using Artificial Neural Network , 2012 .

[54]  Geoffrey E. Hinton,et al.  Acoustic Modeling Using Deep Belief Networks , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[55]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[56]  Mahesh Pal,et al.  Performance evaluation of artificial neural network approaches in forecasting reservoir inflow , 2012 .

[57]  Ozgur Kisi,et al.  Applications of hybrid wavelet–Artificial Intelligence models in hydrology: A review , 2014 .

[58]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[59]  James P. Crutchfield,et al.  Geometry from a Time Series , 1980 .

[60]  Yin Wang,et al.  Dynamic Forecast of Daily Urban Water Consumption Using a Variable-Structure Support Vector Regression Model , 2015 .